Type beta transforming growth factors (TGF beta) are multifunctional regulatory peptides, implicated in myocardial differentiation and hypertrophy by their pattern of expression and by provocative in vitro models. TGF beta signal transduction is thought to involve a heteromeric complex of transmembrane Ser/Thr protein kinases, the type II and type I TGF beta receptor (T beta R-II, T beta R-I). Directional phosphorylation of the Ser-rich TBR-I "GS box" by TbR-II is proposed as a prerequisite for all TGFb effects. By contrast, TBR-I is the component of the receptor complex which is thought to confer cellular responses to TGFb, through interaction with cytoplasmic substrates or other effector molecules. However, these comparative roles postulated for TbR-II and TbR-I are controversial. A counter-hypothesis (the "two-pathway" model), for which support exists in some cell types, proposes instead that TbR-II and TbR-I mediate distinct subsets of TGFb's actions - growth inhibition and gene induction, respectively. Genetic analysis utilizing gain-of-function and loss-of-function mutations provides a powerful tool to move beyond descriptive studies and ascertain the role played by putative regulators of cardiac growth and differentiation. The applicant has recently proven that signal-deficient mutations of TbR-II and TbR-I, in cardiac muscle and non-cardiac cells, each suppress the transcriptional effects of all three mammalian isoforms of TGFb. However, given the potential for dominant-negative receptors to form both homodimers and heterodimers, dominant-negative receptors by themselves cannot distinguish between the "interdependent" and "two-pathway" models. Furthermore, little is known of the cytoplasmic effector molecules that might convey the effects of ligand-activated TGFb receptor. Specific Aims of the present project are: (1) Using site-directed mutagenesis, to replace Ser and Thr residues of the type I receptor GS box with alanine, as a predicted dominant-negative mutation, and with glutamic or aspartic acid, as a predicted gain-of-function mutation. (2) Using interaction cloning in yeast, to identify (and characterize) candidate substrates for the type I receptor kinase domain. (3) Using the Cre/loxP system for organ-specific homologous recombination, to selectively inactive the genes for type I and type II receptor in ventricular myocardium.